Locating the position of an event in acquired digital data at sub-sample spacing

Abstract

A method for characterizing an event in acquired digital data is described where the event has a known shape and a pattern having amplitude and location coefficients is applied to the data for determining a best fit between the data and the pattern as a function of a peak RMS value. The derived RMS value is compared to a threshold value for verifying the existence of the event. The event is characterized as to amplitude and location using the amplitude and location coefficients of the pattern. Such a method is useful in characterizing non-reflective events in acquired optical time domain reflectometry data.

What is claimed is: 
     
       1. A method for characterizing an event in a waveform of acquired digital data where the event has a known shape comprising the steps of: (a) applying a pattern having amplitude and location coefficients over the acquired digital data for determining a best fit between the data and the pattern as a function of a peak RMS value by incrementing the amplitude coefficient of a pattern and positioning the pattern over the acquired data for each amplitude coefficient for determining the RMS value as a function of varying the location coefficient with the location coefficient increment spacing being less than the sample spacing of the digital data;   (b) comparing the peak RMS value to a threshold value for verifying the existence of the event; and   (c) characterizing the event as to amplitude and location using the amplitude and location coefficients of the pattern.   
     
     
       2. The method as recited in claim 1 wherein the comparing step further comprises the step of evaluating the threshold value as a function of the local noise at the event. 
     
     
       3. A method of characterizing a non-reflective event in a waveform of acquired optical time domain reflectometer digital data from a fiber under test where the non-reflective event has a known shape comprising the steps of: (a) applying a pattern having amplitude and location coefficients over the acquired digital data for determining a best fit between the data and the pattern as a function of a peak RMS value by incrementing the amplitude coefficient of a pattern and positioning the pattern over the acquired data for each amplitude coefficient for determining the RMS value as a function of varying the location coefficient with the location coefficient increment spacing being less than the sample spacing of the digital data;   (b) comparing the peak RMS value to a threshold value for verifying the existence of the event; and   (c) characterizing the non-reflective event as to loss and location on the fiber using the amplitude and location coefficients of the pattern.   
     
     
       4. The method as recited in claim 3 wherein the applying step further comprises the step of determining an approximate location for the non-reflective event in the acquired digital data. 
     
     
       5. The method as recited in claim 3 wherein the comparing step further comprises the step of evaluating the threshold value as a function of the local noise at the event. 
     
     
       6. The method as recited in claim 4 wherein the acquired digital data is a function of optical pulses having a predetermined pulse width being launched into the fiber under test and a return optical sign from the fiber is converted to an electrical signal, sampled and stored in the optical time domain reflectometer and wherein the characterizing step further comprises applying a correction value to the location coefficient of the pattern as a function of the amplitude coefficient and the pulse width of the launched optical pulses. 
     
     
       7. A method of characterizing a non-reflective event in acquired optical time domain reflectometry digital data from a fiber under test wherein the non-reflective event has a known shape and optical pulses having a predetermined pulse width are launched into the fiber and a return optical signal from the fiber is converted to an electrical signal, sampled and stored in the optical time domain reflectometer, the steps comprising: (a) determining an approximate location for the non-reflective event in the acquired digital data;   (b) applying a pattern having amplitude and location coefficients over the acquired digital data for determining a best fit between the data and the pattern as a function of a peak RMS value by incrementing the amplitude coefficient of a pattern and positioning the pattern over the acquired data for each amplitude coefficient for determining the RMS value as a function of varying the location coefficient with the location coefficient increment spacing being less than the sample spacing of the digital data;   (c) comparing the peak RMS value to a threshold value for verifying the existence of the event with the threshold value being evaluated as a function of the local noise at the event;   (d) characterizing the non-reflective event as to loss and location on the fiber using the amplitude and location coefficients of the pattern; and   (e) applying a correction value to the location coefficient of the pattern as a function of the amplitude coefficient and the pulse width of the launched optical pulses.